Ketogenic diet produced larger liver improvements during matched weight loss trial

A randomized clinical trial published August 27 in *Cell Metabolism* found that diet composition affected some metabolic responses even when participants lost approximately the same amount of weight. Among adults with obesity, prediabetes and excess liver fat, a very-low-carbohydrate ketogenic diet produced larger improvements in liver fat and hepatic insulin sensitivity than Mediterranean and low-fat, plant-forward diets.

The result is relevant to prevention because prediabetes and metabolic dysfunction-associated steatotic liver disease can precede type 2 diabetes, cirrhosis and cardiovascular complications. It does not establish that ketogenic eating prevents those outcomes, works better for the wider population or is sustainable and safe over years.

A tightly controlled comparison

Researchers randomized 55 adults with obesity, prediabetes and hepatic steatosis to one of three diets. Forty-two completed the intervention. The ketogenic diet supplied about 4% of energy from carbohydrates and 73% from fat. Corresponding proportions were 50% and 35% for the Mediterranean diet and 70% and 15% for the low-fat, plant-forward diet.

Participants received all their food from the research team and met weekly with a dietitian. Calorie intake was adjusted so each participant lost approximately 10% of starting body weight, which generally took four to five months. This design allowed the investigators to compare metabolic effects at similar weight loss while reducing uncertainty about what participants ate.

Endpoints included insulin sensitivity in muscle and liver, intrahepatic triglyceride content, glycated hemoglobin, liver fat production and serial glucose, insulin and triglyceride measurements over 24 hours. These are physiological and laboratory endpoints rather than direct measures of liver failure, diabetes complications, cardiovascular events or survival.

Weight loss helped across all three groups

Muscle insulin sensitivity improved by approximately 50% in each group, supporting weight loss itself as the main driver of that response. All three diets also reduced body fat and improved multiple metabolic measures.

Differences emerged most clearly in the liver. Hepatic insulin sensitivity increased two to three times more in the ketogenic group than in the other groups. Average liver-fat content fell 67% with the ketogenic diet, compared with 45% under both the Mediterranean and plant-forward plans.

Twenty-four-hour glucose concentrations decreased 20% from baseline in the ketogenic group and 8% in each comparator group. Daily insulin concentrations fell 74% with the ketogenic diet, 44% with the Mediterranean diet and 27% with the plant-forward diet.

At the intervention’s end, approximately 50% of ketogenic-diet participants no longer met prediabetes criteria, compared with 29% and 7% in the Mediterranean and plant-forward groups. Those percentages come from small groups and should be treated as exploratory rather than precise estimates of remission probability.

Despite its higher fat content, the ketogenic diet did not produce statistically distinguishable group differences in LDL cholesterol, apolipoprotein B or 24-hour triglyceride concentrations during the trial. That short-term finding does not settle longer-term cardiovascular safety or how individuals with different baseline lipid responses might fare.

What the trial cannot answer

The study’s principal strengths—complete food provision, frequent dietitian contact and a fixed weight-loss target—also limit its real-world applicability. Most people choose and prepare their own food without weekly research support. The trial therefore tested metabolic efficacy under intensive supervision, not long-term feasibility.

The sample was small, 13 randomized participants did not complete the intervention, and follow-up ended after several months. The study was not designed to compare clinical events, liver fibrosis, quality of life, nutritional adequacy or weight maintenance after food provision stopped. It also cannot determine whether the reported advantages persist once carbohydrate intake increases or weight is regained.

The practical conclusion is narrow: among selected adults with obesity, prediabetes and liver fat who achieved matched 10% weight loss under controlled conditions, macronutrient composition influenced several short-term liver and glucose endpoints. Larger and longer trials are needed before those biomarker differences can be translated into claims about durable disease prevention or healthspan.